WO

RK

ING

PA

PE

R

Law versus Economics?

How should insurance intermediaries influence the

insurance demand decision

University of Lüneburg Working Paper Series in Economics

No. 299

June 2013

www.leuphana.de/institute/ivwl/publikationen/working-papers.html

ISSN 1860 - 5508

by

Annika Pape

Law versus Economics?How should insurance intermediaries influence the insurance

demand decision

Annika Pape ∗

June 4, 2013

How should intermediaries influence the insurance demand decision? Theanswer must refer to the interdependence of economic determinants and le-gal duties. Intermediaries potentially guide demand decisions by deliveringobjective information and by considering individuals’ situation and economiccircumstances. The economic theory provides determinants that are essentialfor the insurance demand decision. Undoubtedly, consumers lack informationabout certain variables, and therefore misjudge their demand for insurances.In response to the consumer, an intermediaries’ task is to discover possiblemisjudgments and to provide the correct information. Since the informationin the insurance market is asymmetrically distributed, an insurance agenthas an incentive to behave opportunistically, a tendency that is reinforcedby the remuneration scheme in Germany. In 2007/2008, insurance intermedi-aries became regulated by law. That law states, among other things, the fourbasic obligations of insurance intermediaries and a liability rule to sanctionviolations. In order to interpret and substantiate the legal terms, those haveto match the relevant economic determinants to state the ideal behavior ofan intermediary.

Keywords: insurance, insurance intermediation, advice, liability, InsuranceContract Act

JEL-Classification: G22, D83, D89, K29, K40

∗ Leuphana University Lueneburg - IVWL, Scharnhorststrasse 1,21339 Lueneburg, Germany. E-mail:pape@leuphana.de

1

1 Introduction

Insurances are considered to be complex experience and credence good. Therefore,consumers tend to have difficulties in making informed buying decisions, which mightexplain the preference concerning the distribution channels. In Germany insurance poli-cies are predominantly sold by insurance intermediaries (Insurance Europe, 2010). Therole intermediaries occupy in the insurance market is highlighted by the amount of thepre-tax premiums which added about 7.16% to the German GDP in 2011 (Gesamtver-band der Deutschen Versicherungswirtschaft e. V, 2011, p.56).The insurance market is characterized by imperfections and limited information whichresults in sub-optimal contracting between the insurer and the consumer. The economictask of any intermediary is to partially overcome the resulting market failure. For thatreason, an insurance agent is supposed to match the consumers’ preferences with thesupply of policies by guiding individual insurance demand decisions. Hence, the relevanteconomic determinants display possibilities for the intermediary to intervene and to im-prove the individual choice, ideally. Nevertheless, the economic environment allows foropportunistic behavior by the insurance agent since his knowledge of market conditionsis superior and the existing remuneration scheme supports that tendency. For that rea-son, laws have regulated insurance intermediaries since the implementation of the EUdirective. Legal regulations state obligations an intermediary must fulfill in order to beexempted from liability. From a law and economics perspective, the legal requirementshave to be such that the agent chooses not to behave opportunistically. This can only beachieved if economic determinants are reflected in legal obligations. Furthermore, sincethe liability rule refers to a default, the jurisdiction needs diligence levels as a referencein order to benchmark the intermediaries’ behavior and counseling advices.Recently, the jurisdiction had to deal with complaints about the counseling activities ofintermediaries. Following is a brief look at three cases and the final court decisions:

1. Automobile Insurance - Reference OLG Hamm 20 U 131/091:The consumer (plaintiff) purchased a used RV (recreational vehicle), which waspartially financed by way of credit during a four-year time frame. He demandeda automobile insurance policy from the insurance agent and ended up with apart insurance cover (sum insured: 21.000 EUR) and the mandatory auto liabilityinsurance. Shortly after the policy was taken out, the RV was destroyed in anaccident. It is questionable if the intermediaries’ counseling was faultless sincehe did not give advice that would have resulted in a comprehensive insurancecover in order to secure the credit to a similar level. The court decided that theintermediary did not neglect any legal obligations, inter alia, because automobileinsurances are not considered complex products. Additionally, the consumer hadtaken out that kind of insurance before. Thus, the consumer is conversant withthe differences between comprehensive, full coverage insurance and the liabilitypartial coverage insurance. Also, the fact that the RV was partially financed didnot trigger a duty to mention the option of comprehensive coverage from the courts

1OLG Hamm (Higher Regional Court), court decision 12/04/2009 - 20 U 131/09.

2

point of view. Therefore, the case was dismissed.

2. Residential Building Insurance - Reference LG Ingolstadt 33 O 136/102:In order to update the insurance coverage on her residential building policy, theconsumer (plaintiff) switched to a more current tariff. During the counseling in-terview, the consumer stated her desired for information about the coverage of allwater pipes because of the swimming pool in the backyard. Eventually, one ofthe water pipes in the backyard broke and the water caused 17.000 EUR worth ofdamage, which was not covered by the insurance company. It is questionable if theinsurance agent has to compensate the damage since he stated that all water pipeswere covered, which obviously was not the case. As in the previous example, thecourt decided that the intermediary did not neglect any of his duties, even thoughhis statement concerning the coverage was wrong. However, it cannot be expectedthat an insurance agent extents the coverage beyond the policy conditions whichare assumed to be known by the insurance customer. Additionally, consumersshould know that gutters cannot be covered by insurance because they tend toclog if not maintained regularly. Thus, the case was dismissed.

3. Private Health Insurance - Reference OLG Munich 25 U 3343/113:As a result of a counseling interview with an insurance intermediary, the consumer(plaintiff) canceled his health insurance policy after 25 years in order to update theinsurance conditions. Later, the consumer realized that the new contract changedcertain conditions to the worse. Additionally, it was not possible to transfer theold-age reserves, which were to reduce the premium once the insured turned 65years old. Thus, additional costs resulted from the worsened conditions and theloss of reserves to compensate an increase in premiums. It is questionable if theintermediary obeyed his counseling duties. In this case, the court decided in fa-vor of the plaintiff since the initial contact was established by the intermediary.Additionally, the insurance agent did not correct wrong assumptions about the rel-evant policy changes and he did not discuss the problem of the non-transferabilityof old-age reserves or the resulting disadvantages.

These three examples have the same question in common: was the counseling interviewand the resulting advice at fault for damages or not? The aim of this paper is to statepossibilities for an intermediary to influence the insurance-demand decision as the eco-nomic theory suggests, and to discuss how those economic determinants might be usefulin order to interpret legal duties and to evaluate the faultlessness of counseling advice.As well, the identification and discussion of the influence of economic determinants andtheir possible legal counterparts build Sections 2 and 3. Section 4 states specific the-oretical examples to examine the interaction of law and economics. Finally, Section 5concludes and refers back to the stated court decisions.

2LG Ingolstadt (Regional Court), court decision 12/29/2010 - 33 O 136/10.3OLG Munich (Higher Regional Court), court decision 06/22/2012 - 25 U 3343/11.

3

2 Insurance Demand

The first step towards an answer to the question “How do intermediaries influence thedemand for insurance?” has to be a consideration of the economic theory on insuranceand uncertainty. Research in this field is ongoing since the early 1960s when researchersstarted to relate insurance issues to the economic theory (Louberge, 2000, p.4). Sincethen the expected utility approach remained predominant even though other approachesemerged. In this section the basic model of insurance demand theory is introduced andwill be extended by background risks in order to examine the insurance decisions ofconsumers in the presence of multiple risks.

2.1 The Basics: Insurance Demand Theory

The possibility to buy an insurance policy enables households to reduce or even to pre-vent risks (Zweifel and Eisen, 2012, p.101). In fact, the aversion against risks constitutesthe crux of the insurance demand theory which basically is an application of the expectedutility approach by von Neumann and Morgenstern.In the simplest version of insurance demand theory, supply of insurance is exogenous.This assumption might be crucial if the insurant is a firm instead of a consumer. Whereaslarge firms have the possibility to negotiate customized insurance solutions, the con-sumers must choose from a variety of generally standardized products. Therefore, itseems feasible to stay with the assumption of exogenous supply, for the purpose of thispaper. Considering uncertainty, a household has to form expectations about futurewealth4. A loss (L) might occur with a likelihood (ρ) and yield a reduction of the initialwealth (W0). Assuming only two possible states of the world [W1 = W0 − L,W2 = W0],the consumer expects a final wealth level (WE).

WE = ρ(W1) + (1− ρ)W2 (1)

At this point, the individual attitude towards risk has to be discussed. Only a riskaverse consumer prefers a certain to an uncertain payment, i.e. the utility of the expectedvalue of the lottery or gamble is greater than the expected utility of the lottery (EU(WE))itself.

EU(WE) = ρυ[W1] + (1− ρ)υ[W2] (2)

From equation (1) and (2) follows that, if the expected value of the lottery remainsunchanged, the consumer maximizes the expected utility if the marginal utility of riskywealth is equal in each state.

−(1− ρ)

ρ= −

(1− ρ)

ρ

υ′[W2]

υ′[W1]

υ′[W1] = υ′[W2]

(3)

4The whole section refers to Zweifel and Eisen (2012, pp.71-110). The notation differs slightly.

4

This optimality condition can only be satisfied by equalizing the wealth levels in thetwo states. Graphically, the optimality condition (3) is shown in Figure 1. The indif-ference curve through endowment point A is characterized by a lower level of expectedutility than the indifference curve through point C even though the expected wealth(WE) is identical. Moreover, it becomes obvious that a risk-averse individual is willing

W✁

W✂

•

A

45°

W✂=W✄

W✄

L=I

P(I)

P(I)

*

1W

*

2W

I-P(I)

• C

Figure 1: max. expected utility

to give up a certain amount of money in both states in order to transform a risky situ-ation into a certain situation. By paying a premium (P (I)) the consumer actually buysthe promise that the insurer will provide an indemnification payment I = αL in the caseof a loss. Obviously, the variable (α) specifies the degree of coverage: α = 1 indicatesfull coverage, whereas α < 1 refers to partial cover. The possibility to buy an insurancepolicy changes the final wealth levels as stated in equation (4).

W1 = W0 − L− P (I) + I

W2 = W0 − P (I)(4)

Even though the loss L might not occur, the consumer still has to pay the premium.Therefore, the wealth level in the “no loss” state W2 reduces, too. Despite the insurance,neither state of the world is certain and, therefore, the consumers’ aim is to maximizeexpected utility, given certain probabilities and premiums.

EU = ρυ[W0 − L− P (αL) + αL] + (1− ρ)υ[W0 − P (αL)] (5)

In equation (5) the premium depends on the indemnification payment. Those premi-ums can be further specified in order to distinguish “fair” and “unfair” premiums. Whilefair premiums are equal to the expected losses, unfair premiums exceed them. In orderto demonstrate both cases, the premiums can be stated as P = (1 + λ)ραL. Wheneverthe premium equals the expected losses, the proportional loading (λ) has to be zero.

5

Inserting the stated premium function into equation (5) and optimizing with respect toα yields the optimality condition and implicitly the optimal degree of insurance coverage.

maxαEU = ρυ[W0 − L− (1 + λ)ραL+ αL] + (1− ρ)υ[W0 − (1 + λ)ραL]

υ′[W2]

υ′[W1]=

1− ρ− ρλ

1− ρ+ λ− ρλ

(6)

First, assume a “fair” premium that does not induce any changes to the level ofexpected wealth and yields expected profits of zero to the insurance company. As stated,in this case the loading λ equals zero. The optimal condition can be derived fromEquation (6) and again states that it is optimal to equalize the marginal utilities of riskywealth, υ′[W1] = υ′[W2]. The only way to accomplish that condition is to fully coverthe loss, implying α = 1. After considering the optimal insurance demand assuming apremium equal to expected losses, the case of unfair premiums shall be discussed next.In this case, the proportional loading λ has to be positive. From Equation (6), it followsthat in this case the marginal utility of risky wealth in the loss state has to be greaterthan the marginal utility in the state without a loss. Because of the shape of the utilityfunction of a risk averse individual, it has to be the case that the level of wealth withouta loss exceeds the final wealth level in the loss state, W2 > W1. Therefore, it is notoptimal to buy comprehensive full coverage but to insure partially which forces α to beless than 1.

2.2 Multiple Risks

In the previous section, insurance demand was analyzed while focusing only on onerisk a consumer is confronted with. Furthermore, the basic model solely accounts forinsurable risks under complete market conditions. Undoubtedly, consumers face severalrisks at the same time - which might as well be interdependent. Doherty and Schlesinger(1983) state that for an optimal insurance demand, the correlation between insurable anduninsurable risks matters. They introduce an uninsurable background risk to accountfor some incompleteness in the insurance market. Considering this additional risk, theconsumer is confronted with four possible states of the world: 1. the initial wealthlevels remains unchanged; 2. the insurable loss (L) occurs and reduces the wealth levelaccordingly; 3. only the uninsurable loss (N) happens, or 4. both incidents take place. Asingle loss arises with either probability ρL or ρN . Considering ρL and ρN , it is possibleto assign a probability of occurrence (ρ1 − ρ4) to each state. Comparing this modelto the insurance demand theory stated in the previous section, it is obvious that theexpected utility function has to be adjusted for the additional states. Furthermore, the

6

premium function has to be adapted slightly by comparison with the previous sectionand becomes P = (1 + λ)ρLαL.

EU = ρ1υ[W0 − (1 + λ)ραL]

+ ρ2υ[W0 − (1 + λ)ραL− (1− α)L]

+ ρ3υ[W0 − (1 + λ)ραL−N ]

+ ρ4υ[W0 − (1 + λ)ραL− (1− α)L−N ]

(7)

The evaluation of Equation (7) at α = 1 yields Equation (8).

dEU

dα|α=1 = ρLL[ρN(1 + λ)− ρN |L] · [υ

′[W1]− υ′[W3]− υ′[1]λ] (8)

At this point, it is possible to determine the optimal degree of coverage while con-sidering different cases. Assuming a fair premium (λ = 1) and independence of thelosses, i.e. ρN = ρN |L, Doherty and Schlesinger (1983) state that it is optimal to optfor full coverage. Whenever the losses are correlated, either positively (ρN < ρN |L) ornegatively (ρN > ρN |L), it is optimal to demand more than full coverage, or even onlypartial coverage respectively. To demand more than full coverage in the event of a pos-itive correlation between L and N , becomes plausible if one considers a compensationof the background risk N . In the case of an unfair premium (λ > 1) partial coverageis optimal if there is no correlation or a negative correlation. Whereas in the event ofpositive correlation, the optimal degree of coverage is indeterminate and depends onmagnitude of the loading λ (Doherty and Schlesinger, 1983), or on the size of the loss N(Zweifel and Eisen, 2012, p. 99). Fei and Schlesinger (2008) extend the outlined modelon multiple risks by state dependency. They show that if N is larger in a state of theworld with an insurable loss, the consumer demands more than full coverage, even if thepremium is fair. On the contrary, if N is larger in a state of the world when no insurableloss occurs, the consumer demands only partial coverage. A precautionary motive servesas an explanation for that behavior in each case. In the first case, the precautionarymotive yields a compensation of the uninsurable loss by over-insuring the other loss. Inthe second case, the consumers’ precaution results in partial coverage in order to reducethe premium and, thus, to be able to fund the background risk.

2.3 Discussion - Determinants of Insurance Demand

The previous section outlined the insurance demand theory by simply assuming that allrelevant information is known. As a result, the determinants of insurance demand canbe derived. There is: 1. the degree of risk aversion, 2. the premium, 3. the wealthlevel, and 4. the correlation with uninsurable background risks. Considering the degreeof risk aversion, the dislike of risks leads to the demand of insurances in the first place.It is plausible that a higher risk aversion affects the degree of coverage whenever thisdiffers from a comprehensive payment. Moreover, even if the premium is loaded, theinsurance demand increases whenever the degree of risk aversion rises. Additionally,

7

Mossin (1968) found that in cases of constant and decreasing absolute risk aversion, theconsumers tend to reduce the insurance demand as a reaction of a loaded premium.Since insurers have to cope with administrative costs, anticipate moral hazard, and tohave some liquidity reserves, it is quite feasible that the premium always differs from afair premium. Thus, the theory predicts that partial coverage would be optimal. Thecoverage reduction can be established by either a deductible, or alternatively a coin-surance. But, the previous section also states that one might not be able to generalizethose findings since the correlation with an uninsurable background risk is possibly con-tradicting. In addition to the stated determinants, the consumer also has to have a goodidea about possible states of the world and the set of potential actions, the probabilityof the occurrence of a loss, and financial consequences in each state in order to solve thedecision-making problem (Zweifel and Eisen, 2012, p.75).

Consumers sometimes behave other than the classical theory predicts and for examplebuy too much insurance or, stated differently, buy insurances with low or no deductibleseven if the premium is loaded (Shapira and Venezia, 2008). Such anomalies mightarise because the consumers make mistakes (Schwarcz, 2010). Among mistakes are anoverestimate of the probability of a damage, or even uncertainties about the effect of aloss event on the final wealth level (Mossin, 1968). Schwarcz (2010) offers an alternativeexplanation whereupon consumers value an insurance more than the expected utilityfunction is able to prognosticate, so their behavior is not a result of limited informationor knowledge. If the latter hypotheses is true, it should be impossible for an insuranceagent to influence the insurance demand at all. But, Schwarcz (2010) analyzes differentdemand anomalies such as insurance demand for small financial risks, or preferencesfor insurances with small deductibles, and he finds that people simply make mistakeswhile taking out insurance policies. Additionally, consumers can hardly cope with themultitude of competing insurance companies and their differentiated products. Bothaspects, the tendency to estimate probabilities of biased occurrences, and the fact thatpeople cannot overlook the insurance market and therefore use heuristics to determinetheir demand for insurance, cause sub-optimal decisions (Hofer, 2008; Jaspersen andAseervatham, 2012; Schwarcz, 2010). Since it seems plausible to assume that the demandfor insurance reflects an uninformed choice that might additionally be driven by cognitivelimitations, it has to be possible for an expert to intervene and, thus, to enable a betterchoice.

3 Intermediary: Duties and Responsibilities

Whereas the last section dealt with the basic insurance demand theory, outlined theknowledge and the information a consumer ideally needs to solve the decision-makingproblem and discovered the existence of demand anomalies, this section focuses on thetasks an intermediary has to cope with.

8

3.1 Influence upon Economic Determinants

Consumers might make mistakes while taking out insurance policies. This fact leadsdirectly to the tasks an insurance agent should perform from an economic point ofview. As pointed out, a consumer has to have a good idea of the probability of theloss occurrence. If this estimation is biased, the insurance demand decision concerningthe coverage will be sub-optimal. Let the objective probability of a loss be ρ, but thesubjective estimation of the loss be q with q > ρ, i.e. an overestimation of the lossprobability. Additionally, assume a loaded premium so that a partial coverage becomesoptimal. In a case with both, a loaded premium and an overestimation, the approachto maximize the expected utility changes from equation (6) to equation (9).

maxαEU = qυ[W0 − L− (1 + λ)ραL+ αL] + (1− q)υ[W0 − (1 + λ)ραL] (9)

Equation (9) states that the overestimation of the loss occurrence does not affect thepremium calculation by the insurer, because the mistake is assumed to happen on thepart of the consumer. Figure (2) illustrates the effect of the overestimation. First,

W�

W☎

•

A

45°

W☎=W

✆

•

E •

D

W�

Figure 2: Optimal coverage with a loaded premium and overestimation of the loss prob-abilities

Figure (2) shows that it is optimal to choose D whenever the premium is loaded and theprobability of loss occurrence is still estimated correctly. But, as soon as the consumeroverestimates the loss occurrence, the expected utility indifference curve becomes flatter,which results in E as the optimal choice. Nevertheless, the dashed indifference curveis the result of a mistake. Hence, by choosing E instead of D the consumers’ expectedutility level decreases because the insurance coverage is too high.Equation (10) states the same result formally. Compared with a situation in which thepremium is loaded, but the estimation of the loss probability is correct, the fraction onthe right hand side of the equation increases because ρ < q. The increase of the fraction

9

is equivalent to a convergence of wealth levels in the different states which can only beaccomplished by a rise in α.

υ′[W2]

υ′[W1]=

q − qρ− qρλ

ρ− qρ+ λρ− qρλ(10)

Since any divergence from the optimal α results in a loss of expected utility, the inter-mediaries’ task in this case should be to discover the objective loss probabilities underconsideration of the individuals’ situation.The stated example shows quite obviously, that an intermediary might be able to fixthe mistakes since he is an expert, and therefore his choices should be superior to theconsumers’ considerations. The insurance demand is driven by several determinants aspointed out in Section 2. From an economic perspective almost all those determinantsmight be the cause of mistakes, and thus establish opportunities to act for an insuranceagent. Consumers might misjudge the amount of a loss, or they do, or do not considerthe existence of uninsurable background risks. Additionally, consumers might not evenbe aware of some risks, and therefore they do not take all relevant states of the worldinto consideration during the decision-making process. Undoubtedly, since consumerscan hardly cope with the multitude of insurance products, they might not take all pos-sible actions into account. Importantly, preventive effort can reduce or avoid losses andshould be considered as an alternative to insurance products or even as a complementaryaction.From an economic point of view, the question of how insurance intermediaries shouldinfluence the insurance demand decision can be answered when focusing on the deter-minants of insurance decisions and possible mistakes. Nevertheless, the degree of riskaversion represents a factor that is neither known by the intermediary, nor can it be influ-enced. This insight leads to the statement that an intermediary can deliver informationand can even correct mistakes, but the final evaluation of the product alternatives andthe final choice can only be performed by the consumer himself. Thus, the intermedi-ary can only provide support by counseling and delivering information, but he cannotentirely solve the decision-making problem on the consumers’ behalf.

3.2 Legal Obligations

In contrast to the previous part, this subsection will focus on the legal instead of onthe economic point of view because the working conditions of intermediaries changedas a result of market interventions by the European Union. Initially, the supply ofinsurance products increased in the 90’s as a result of the deregulation of the Europeanmarket in 1994. Moreover, the products themselves became more heterogeneous sincethe preventive control of the general insurance conditions by the regulation authoritieswas abolished in the cause of the deregulation process (Ihle, 2006, p.52). Even thoughthe deregulation stimulated the competition in the field of insurances, the consumerprotection suffered losses (Hofer, 2008, p.93). In order to re-strengthen the consumerprotection, the European Union issued a directive in 2001 that regulates the insuranceintermediaries and their counseling activities by law rather strictly.

10

The German Insurance Contract Act (VVG)5 lists several duties, which an insuranceintermediary must fulfill. First, the intermediary has to base his advice upon a sufficientnumber of products that are suitable for the individual situation of a consumer withregard to objective criteria, § 60 VVG. For the number of evaluated policies to be“sufficient”, a broker has to refer to the entire market, whereas other agents - tiedagents in exclusivity or agents that cooperate only with a couple of insurers - have tosolely consider products of their co-contractors (Dorner, 2010, § 60 recitals 3,14,15).If the advice is based upon selected companies or products only, the consumer has tobe notified (Michaelis, 2010, p.67). Also, the Insurance Contract Act states that theinsurance agent has to ask the consumer about his wishes and needs, § 61 VVG, in orderto obtain all relevant information. Whereas the wishes are of a rather subjective nature,the intermediary is supposed to focus on the objective needs of a person, and if necessary,to bring both aspects in line (Michaelis, 2010, p.86). Additionally, if obviously wrongassumptions about the needs or the details of an insurance policy becomes known, anintermediary has to fix those mistakes (Dorner, 2010, §61 recitals 25,26). Furthermore,§ 61 VVG states an obligation to give advice. The intermediary has to discuss howand why he selected the products, and also, which of the alternatives constitutes thebest choice given the circumstances of the single consumer. To sum up, the InsuranceContract Act assigns basically four duties an intermediary has to meet: 1. provideinformation based upon a sufficient number of alternatives; 2. self-dependently inquirerelevant factors; 3. provide advice and the reasons for the choice, and 4. the duty to keeprecords. In contrast to § 347 para. 1 HGB6 which states the due care and diligence of aprudent businessman referring to commercial transactions, the VVG further specifies thecriterion for diligence. Additionally, the Insurance Contract Act refers to the individualsituation of a consumer. Therefore, the applicable diligence level is dependent on theparticular case (Zinnert, 2010, p.53), but the intermediary still has to obey the diligenceof a prudent businessman. Furthermore, the Insurance Contract Act states a liabilityrule in § 63 VVG: the intermediary is liable for any damages, whenever he neglects hisduties. Of course, the breach of duty and the resulting damages have to be causallydetermined in order to trigger the liability, and thus, yield a compensation.

3.3 Discussion: Law and Economics

First, intermediaries can influence the insurance demand decision of consumers by mod-ifying the economic determinants, or by simply adding relevant information. Second,the Insurance Contract Act assigns four legal duties that an intermediary has to meet;however, the liability rule which is supposed to ensure that the insurance agent performshis task properly, refers to the legal duties instead of the economic variables. Since the

5Insurance Contract Act (Versicherungsvertragsgesetz (VVG)) November 23, 2007 (Federal LawGazette Part I, page 2631), becoming effective on 01/01/2008; last changes 07/27/2011 (FederalLaw Gazette Part I, page 1600) with effect from 08/04/2011.

6German Commercial Code (Handelsgesetzbuch (HGB)) May 10, 1897 (Federal Law Gazette Part III,No. 4100-1), becoming effective on 01/01/1900; last changes 12/20/2012 (Federal Law Gazette PartI, page 2751).

11

economic determinants are more specific than the legal duties, the variables can be usedto substantiate those obligations. Hence, it should be possible to match each economicvariable to a legal duty in order to combine law with economics.

Legal Duties Economic Determinants

Provide information about possible actions, i.e. possible policiesalternatives premiumInquiry of relevant factors objective possibility of loss occurrence

potential amount of the losspossible states of the world

Advice correlation with uninsurable background riskspreventive effort to reduce lossessituational: selection of alternatives

Documentation ———— degree of risk aversion

Table 1: Economic Determinants / Legal Duties

Table 1 shows that the economically relevant variables have legal counterparts. Forexample, the duty to base the advice upon a sufficient number of potential alternativesmatches the need of knowing the whole set of possible actions in order to solve theeconomic decision-making problem. The necessity to evaluate the monetary consequencein each relevant state of the world, finds its counterpart in the duty to inquire therelevant factors. Of course, those stated duties are preconditions to the duty to giveadvice, but since the advice should be well founded, this duty is more far-reaching.Thus, the correlation with uninsurable background risks, as well as the supplementaryconsideration of preventive effort and precaution, are part of the duty to give advice.However, the degree of risk aversion as an economic determinant can hardly be allocatedto one of the legal duties because it is subjective and unquantifiable. Also, there is noeconomic need for any documentation even though that legal duty helps to provideevidence for any counseling or miscounseling.As far as the liability is concerned, § 63 VVG states that whenever an insurance agentneglects any of his duties, he has to compensate the resulting losses. From an economicpoint of view, it is the utility level that matters for compensation (Cooter and Ulen, 2004,p.312). In the case of uncertainty, it should be the level of expected utility. Considerthe previously introduced example in which the consumer overestimated the probabilityof a loss and ended up with too much insurance. In this case the intermediaries’ duty isto inquire all relevant factors objectively and to re-estimate the loss probability. Figure3 represents the indifference curves that correspond to the objective loss probabilitiesρ. Those indifference curves imply the expected utility levels that arise from the wealthlevels in points D and E. A consumer who overestimates the loss occurrence will chooseE, and is therefore worse off than in the optimal point D. Thus, an insurance agentwho does not correct the initial estimation has to pay a compensation that enables the

12

consumer to obtain the same expected utility level which he would have had by choosingD. In the depicted case, the consumer pays a premium that is too high. To recover theutility level, the payment would have to result in a higher wealth level in the loss statethan the consumer actually experiences in E. On the contrary, the premium, as wellas wealth levels, are accounted for in monetary terms. Therefore, it would be a feasiblecompensation to force the insurance agent to refund the overpaid premium ∆P (I). Inreturn, the consumer has to give up an amount ∆W1 to reduce the indemnificationpayment. As a result, D will be restored.

W✝

W✞

• A

45°

W✞=W✟

W✟

1W∆

• D

W✝ P(I)∆

•E

Figure 3: Compensation in a case of an overestimation of loss probabilities

4 Examples

The last section discusses the interdependencies of legal obligations and economic vari-ables as well as possibilities to compensate consumers. So far, the analysis bases upongeneral examples without any link to specific insurance policies. This section, on the con-trary, will discuss the ideal behavior of intermediaries on the basis of selected prevalentinsurances.

4.1 Household Contents Insurance

A household contents insurance covers personal belongings against losses and ranksamong the predominantly taken out insurance policies in Germany (Allensbacher Wer-betrager-Analyse, 2011, p.55). It compensates the replacement value of the householdelements. Therefore, the insured sum is based upon the amount which is needed toreplace the whole contents. Despite the wide distribution of that kind of insurance cov-erage, consumers tend to underestimate the amount of a possible loss. That mistakeis of serious consequence. An insurer who discovers the underinsurance of a client will

13

often settle claims only proportionally even if the sum insured is not reached.This case can be analyzed within the economic model of insurance demand: First, thetotal wealth of a consumer has to be decomposed into financial assets on the one hand,and into physical property on the other hand, whereas the latter contains the insurablehousehold elements. In the event of a loss, it is the value of the physical property thatmatters. Again, the starting point will be a situation without any mistakes. In Figure4, the endowment point A depicts the total wealth level W0 on the abscissa and thewealth level of all assets but the value of the household contents on the ordinate. Ata given loss probability ρ, it is optimal to choose wealth levels corresponding to pointD, as represented. Considering a fair premium, it is optimal to take out full insurancecoverage. However, a consumer who underestimates the value of his personal belongings

W✠

W✡

•

45°

W☛

•

• E •

P(I☞)

•

F

D

W✌

A‘

W‘☛

A

P(I✍) P(I✍)

Figure 4: Underestimation of a potential loss

misjudges the amount of the physical property. Hence, the total wealth is underesti-mated. Consequently, the insurance line shifts inward and the consumer considers A′

instead of A as the relevant endowment point. Also, the insurer calculates the premiumaccording to the wrongly estimated and also requested loss amount. Finally, the con-sumer considers himself to face a wealth level in each state according to point E aswell as a corresponding level of expected utility. But, the premium payment (P (IE))actually only yields point F in Figure 4, hence, a situation in which the consumer isunderinsured. Since the resulting α in F is too low compared with the full coverage, theconsumer faces a loss in terms of expected utility as shown.How should an intermediary influence the stated insurance demand decision? Whatactions and counseling activities can be expected? Since the underestimation resultsfrom a wrong evaluation of the existing values, an insurance agent is expected to eitherdetermine the approximately right value of the personal belongings by precise and spe-cific questions or to gather the information by himself during a visit to the consumer’shouse. The alternative is to agree upon a waiver of underinsurance. In this case, the

14

calculations are based upon a predetermined amount insured per square meter 7. Bothalternatives are suitable actions in order to avoid the mistake of an underestimation ofthe potential loss. An intermediary who fails to verify the potential loss amount, and, ifnecessary, fails to correct the initial estimations, neglects his duty to inquire all relevantinformation. Also, by applying the diligence criterion for businessman, an intermediaryacts carelessly if he does not double-check the possibility of an underinsurance becausehe should be aware of those commonly made mistakes. Thus, the intermediary is liablefor the resulting damages.

4.2 Occupational Disability Insurance

A disability insurance maintains the livelihood in the case of an inability to work as aresult of an accident or illness. In 2011, one in four employed persons held a disabilityinsurance policy (Allensbacher Werbetrager-Analyse, 2011, p.55) whereas the remaining75% insure that risk either differently (such as via saving accounts), or not at all. Infact, consumer protection organizations state that consumers tend to underestimate therisk of an occupational disability, which is in line with the findings of representativestudies on behalf of large insurance companies (Continentale Lebensversicherung AG,2011; Birkner, 2012).In contrast to the previous example, an underestimation of the loss probability willresult in too little insurance, if one follows the economic analysis. Figure 5 depictsgraphically the effects of the stated underestimation. Still, the points A and D repre-sent the endowment point respectively the ideal insurance demand decision given theeconomic determinants. An underestimation will increase the absolute value of the slopeof the indifference curve. The case of a loss becomes more unlikely, and therefore theconsumer is not willing to give up as much wealth in the no-loss state to gain wealth inthe loss state compared with a situation in which the estimation is unbiased. Given thewrong estimation, point E in Figure 5 is the optimal choice. But, as the α is too lowcompared to the true optimum, the consumer does not maximize the expected utility inE. In the case of occupational disability insurance policies, an insurance agent shouldinform the consumer about the actual probability of loss occurrence since the underesti-mation seems to be a common mistake. However, Birkner (2012) states that consumersmisconceive the causes of the inabilities to work. Whereas mental disorders constitutethe main cause of occupational disabilities (41%, Deutsche Rentenversicherung (2012)),consumers consider problems with the musculoskeletal system to be predominant, fol-lowed by accidents. Hence, consumers underestimate their individual probability of therisk mainly because they misjudge the causes. Thus, consumers might find it reasonableto substitute an occupational disability insurance by an accident insurance since thepremium for the latter usually is lower. From an economic point of view, the consumersnot only misjudge the probabilities of the possible states of the world, they are alsonot aware of them. Additionally, consumers lack knowledge about possible substitutesrespectively about the loss of coverage which comes along with a assumed alternative. In

7At the moment, this amount is around 650 EUR per square meter.

15

W✎

W✏

•

A

45°

W✏=W

✑

*

1W

*

2W

•

D •

E

W✎

Figure 5: Underestimation of the loss probability

this case, the insurance agent has to inform the consumer about different types of moreor less substituable contracts (duty to inform), and about potential misjudgements thathave to be corrected based upon objective facts (duty to inquire all relevant factors).In contrast to the previous example, the case of the occupational disability insuranceshows that the consumer not only faces an insurance demand decision in terms of thedegree of coverage, i.e. the α, but also faces a selection problem as far as differentiatedbut still substitutable products are concerned. Only if the consumer has informationabout the monetary consequences in each possible state with regard to perhaps slightlydifferentiated products and the corresponding loss probabilities, he is able to evaluatethe expected utilities and finally pick the contract that is optimal given the individualdegree of risk aversion. Thus, the intermediaries’ task is to enable that choice by supply-ing objective and complete information. Stated differently, an insurance intermediarywho provides selected information only, and thus indirectly pushes the consumer to takeout a certain policy is liable for the given advice and potentially has to compensate aresulting loss.

4.3 Legal Protection Insurance

The legal protection insurance basically covers the costs and expenses of legal proceed-ings and can be found in about 25% of German households (Allensbacher Werbetrager-Analyse, 2011, p.55). A legal protection insurance policy can refer to single fields of law,but it is also possible to choose a combination of different fields within a single contract.Usually, a legal protection insurance pays for legal proceedings after a waiting period ofthree months. However, each insurer provides a list of cases that are not insured, forexample the risk in connection with building projects including its funding or disputesconcerning family affairs.

16

In order to solve the insurance demand decision problem, a consumer needs to knowwhich fields of law are relevant for his single situation. Stated economically, the con-sumer has to be aware of all potentially relevant states of the world and their probabilitiesof occurrence. Next, excluded cases of each insurance policy have to be systematicallyevaluated concerning their relevance for the individual. Additionally, the consumer hasto compare the premiums of single policies that refer to single fields of law to those prod-ucts which already combine different aspects. It might be the case that a combination ofpolicies is cheaper than single components or vice versa8 which should be noticed sincethe premium directly affects the consumers’ expected utility as shown in Figure 6. Asthe loading increases, the insurance line flattents out. As a result of the higher premium,it becomes optimal to buy less coverage than point D. An insurance agent has to point

W✒

W✓

•

A

45°

W✓=W

✔

W✔

*1W

*

2W

•

D

•

F

P(I)

0.9λ =

0.2λ =

• E

Figure 6: Differences in premiums

out possible policies in the field of legal protection insurances and has to inform aboutdifferences in the premiums of identical products which might result from combinations.Additionally, he has to discover which states of the world seem relevant for the indi-vidual consumer in order to prevent an overestimation of the specific loss probabilities.If the potential insurant neither owns a car, nor is in possession of a driver’s license,legal protection concerning traffic issues might be neglected because a legal dispute isunlikely. An other example might be that a consumer who lives in his own house, doesnot primarily need legal protection in the case of conflicts between tenant and owner.

8For example: Combinations of legal protection in the fields private and traffic. VGH Insurance:The combination is 7.5% cheaper than single policies (www.vgh.de). Advocard Insurance: Thecombination is 13.3% more expensive than two single policies (www.advocard.de).

17

4.4 Discussion

Consumers make mistakes while taking out insurance policies. Shapira and Venezia(2008) basically state that consumers buy too much insurance and also Mossin (1968)argues that individuals take out full coverage instead of partial coverage. Even thoughone can imagine such mistakes result from overestimations of loss probabilities, or evenan overestimation of the potential loss, the given examples state also situations in whichthe insurance coverage is too low. However, the effect is comparable: an other than theoptimal choice results in utility losses.The existing remuneration scheme in Germany rewards the sale of insurance policies,whereas the amount of that payment might also depend on the premium rate paid bythe consumer (Beenken, 2011). On the one hand, if only the commissions matter forinsurance agents, underinsurance should not be a problem since the agent always hasan incentive to sell more insurance than the consumer initially demands. Neverthe-less, to analyze a consumer’s situation requires effort and is therefore costly from anintermediaries’ point of view. Without the possibility to monitor the intermediaries’work, the incentive to spend only little effort on counseling activities is high as pre-dicted by the principal-agent theory. On the other hand, there is no incentive to sell lessthan the demanded coverage in the case of overinsurance. Additionally, a third aspectemerges. Since some of the products are differentiated but still partially substitutable,the consumer faces a selection problem. Consider for instance term versus endowmentlife insurances.The design of the liability rule has to consider both overinsurance and underinsuranceas well as selection issues. Therefore, the diligence level of each duty has to be setappropriately. Neither an overload of diligence, nor too low requirements are efficient.On the one hand, information that is easily available has to be collected and consideredwhile giving advice. On the other hand, the information should only be acquired andkept in mind by the intermediary if the consumer cannot collect and consider that in-formation cheaper. However, the insurance agent has expert knowledge and therefore isusually considered to collect information at lower costs than the consumer. That neces-sary expertise, on the contrary, results in a rather weak self-information responsibilityon consumers (Curti, 2005, p.211), and a more severe obligation on the intermediaries’part. The examples show that the complexity and, thus, the demand for informationand advice differ according to the insurance policy which has to be considered whendefining and applying a certain diligence level.

5 Conclusion

How should intermediaries influence the insurance demand decision? The answer to thatquestion has to refer to the interdependency of economic determinants and legal duties.Intermediaries have the possibility to guide the demand decision by delivering objectiveinformation on the one hand, and by considering the individuals’ situation and economiccircumstances on the other hand. The economic theory provides determinants that areessential for the insurance demand decision. Undoubtedly, consumers lack information

18

about certain variables and, therefore, misjudge their demand for insurances. So, an in-termediaries’ task is to discover that misjudgment and to provide the right information.Since the information in the insurance market is asymmetrically distributed, an insur-ance agent has an incentive to behave opportunistically - a tendency that is reinforcedby the remuneration scheme in Germany.Insurance intermediaries are regulated by law. Among other things, the law states basi-cally four obligations and a liability rule to sanction any violations. In order to interpretand substantiate the legal terms, they have to match the relevant economic determinantsto state the ideal behavior of an intermediary. The theoretical examples show that thecounseling activity necessarily varies between different products. Whereas, an interme-diary has to fix overestimations or underestimations of probabilities, or potential lossamounts in some cases, he has to deal with selection problems in others. The liabilityrule is supposed to ensure that the intermediary provides the right information in eachof the cases at the individual level. Since the diligence standard is at least comparableto the one a prudent businessman has to meet, an insurance agent is liable for actingcarelessly which assumes that a risk was noticed, but still ignored at the time the coun-seling took place. To sum up, the diligence standard formulated by the law has to beapplied to the specific economic determinants of insurance demand which are within theintermediaries’ sphere of influence. For this purpose, the combination of legal obligationsand economic variable can be used as a checklist to cover all relevant aspects of an idealcounseling interview. Now, let’s return to the introductory cases and court decisionsand take all economic factors and their legal counterparts into consideration.The first case dealt with an automobile insurance for a used RV that was partiallyfinanced by credit. The question is how an ideal counseling interview could have influ-enced the insurance demand decision? First, the intermediary has to base the adviceupon a sufficient number of contract alternatives to provide the consumer with a set ofpossible actions. In this case, the intermediary should inform about the different alter-natives, i.e. part vs. comprehensive insurance coverage as well as the resulting premiumdifferentials. Being an expert, the intermediary knows the relevant differences and caneasily determine the premiums. Second, all relevant factors have to be inquired. Sincethe RV is partially financed, the need for a comprehensive coverage should be put intoquestion as well as the objective amount in the case of a constructive total loss since theinsurance should reduce the financial risk of the credit repayment in the loss state. Therelevant information was available because the consumer mentioned the credit duringthe counseling interview. Depending on the individual financial situation of the con-sumer - the new RV was paid for by turning in an old car and an old RV in additionto the downpayment - the intermediary should have at least offered the comprehensivefull coverage. As far as the documentation is concerned, it has to be stated that in thecase of financing via credit, a comprehensive coverage reduces the financial risk of loanrepayment in several events whereas a part insurance covers only certain occurrences.In the second case, a residential building insurance, a water pipe burst, but the resultingdamage was not part of the insurance coverage. Which information should have ideallybeen provided by the intermediary? In this case the consumer was interested in anupdate of the insurance conditions and an adjustment of the sum insured. Thus, the

19

insurance agent has to inform about all possible contract alternatives that are withinhis portfolio and can serve as an alternative to the current contract. Also, the main dif-ferences should be stated. Since the consumer explicitly mentioned the water pipes andthe swimming pool in the backyard, the intermediary has to refer to the correspondingpolicy details and has to fix the incorrect assumptions about the enclosure of certainwater pipes. Even though the pipes might have not been part of either policy,the oldand the new one, the consumer obviously could not distinguish between fresh water anddrain pipes. Whereas the fresh water pipes are included, the drain pipes are not. Thisdifferentiation should have been part of the counseling interview as well as the note thatthe new insurance conditions do not differ from the old ones concerning this aspect.The advice should have been to switch from the current to the new conditions as longas the consumer is better off. With regard to the drain pipes, the intermediary hasto state that, so far, those are not included. That information gives the consumer theopportunity to take reasonable precautions. In fact, some insurers offer supplementaryinsurances concerning drain pipes and the intermediary should provide or at least men-tion those offers. The documentation should elucidate about enclosures and exclusionsfrom insurance coverage as well as the differences of the two tariffs. If an additionalcontract was part of the counseling interview, the documentation has to refer to thatinformation as well, even if the consumer did not accept it.Finally, in the last case, the consumer not only switched to a new tariff but changed insur-ance companies as well. None of the accumulated old-age reserves could be transferred.Additionally, the patient-centered care differed in some aspects. The intermediary estab-lished the initial contact by offering an alternative to the existing policy. Nevertheless,he still owes advice that is based upon a sufficient number of contracts. Therefore, heshould state the advantages and disadvantages of every alternative in order to presentthe whole set of possible actions. Concerning the wishes and needs of the customer,the new policy conditions should at least be equal to the existing ones (chief physician,single room, private ward). In the case of a setback, the insurance agent has to eval-uate if the consumer accepts this - perhaps in exchange for a different, higher valuedbenefit. Additionally, the intermediary has to calculate the transferable amount of theold-age reserves and has to inform about resulting consequences of the non-transferablepart. In this case, the lower premium of the new policy cannot offset the loss of theold-age reserves and the resulting consequences for the premium development once theconsumer retires. Thus, the insurance intermediary should advise against any change.The documentation has to state that any change will result in a financial loss.Law versus economics? Actually, the paper demonstrates that both aspects complementone another. Whereas the theory of insurance economics focuses on relevant and neces-sary determinants that have to be known in order to solve the decision-making problem,the insurance law provides the legal instrument to force the parties to reveal privateinformation.Still, some aspects remain unanswered. As soon as insurance policies are assumed tobe heterogeneous and perhaps represent substitutes, the decision-making problem nolonger refers to the optimal determination of coverage only, but also states a selectionproblem a consumer has to deal with. Another aspect refers to the provided information.

20

Recall the example of the household contents insurance. A consumer underestimates hisphysical capital. The intermediaries’ advice is to agree upon a waiver of underinsurancewhich yields a way too high insurance amount, and as a result to a premium that is toocostly. The intermediary solved the problem of underinsurance, but at the same time, adifferent mistake happened. Was the advice right or wrong?

References

Allensbacher Werbetrager-Analyse (2011), Versicherungsschutz der Haushalte2010/2011, in Gesamtverband der Deutschen Versicherungswirtschaft e. V, ed.,‘Die deutsche Versicherungswirtschaft. Jahrbuch’, Berlin, p. 55.

Beenken, M. (2011), Provisionen und Courtagen: was die Versicherer ihren Vermittlernzahlen, VersicherungsJournal Verlag GmbH, Ahrensburg.

Birkner, G. (2012), Berufsunfahigkeit 2012 - das unterschatzte Risiko: Wie BerufstatigeRisiken absichern - eine Studie ; Risiko & Vorsorge im Fokus, FAZ-Inst. furManagement-, Markt- und Medieninformationen, Frankfurt and M.

Continentale Lebensversicherung AG (2011), ‘Continentale-Studie zur Beruf-sunfahigkeit: Berufsunfahigkeit - das unterschatzte Risiko’.

Cooter, R. and Ulen, T. (2004), Law and economics, The Addison-Wesley series ineconomics, 4. ed. internat. ed. edn, Pearson Addison-Wesley, Boston Mass.

Curti, H. (2005), Vertragsvermittler und opportunistisches Verhalten: EinLosungsvorschlag aus Sicht der okonomischen Analyse des Rechts, P. Lang,Frankfurt am Main and New York.

Deutsche Rentenversicherung (2012), ‘Rentenversicherung in Zeitreihen’.

Doherty, N. A. and Schlesinger, H. (1983), ‘Optimal insurance in incomplete markets’,Journal of Political Economy 91(6), 1045–1054.

Dorner, H. (2010), §§ 59–73, in E. R. Prolss and A. Martin, eds, ‘Versicherungsvertrags-gesetz’, Vol. 14 of Beck’sche Kurz-Kommentare, Beck, Munchen.

Fei, W. and Schlesinger, H. (2008), ‘Precautionary Insurance Demand With State-Dependent Background Risk’, Journal of Risk & Insurance 75(1), 1–16.

Gesamtverband der Deutschen Versicherungswirtschaft e. V (2011), ‘Die deutsche Ver-sicherungswirtschaft. Jahrbuch’.

Hofer, C. (2008), Produktauswahl in der privaten Krankenversicherung aus Kundensicht:Auswahlprobleme und Verfahren zur Unterstutzung der Auswahlentscheidung, VVW,Karlsruhe.

21

Ihle, J. (2006), Der Informationsschutz des Versicherungsnehmers, Hamburg, Bonn.

Insurance Europe (2010), ‘Cea statistics no 40’. Data 1999-2008.

Jaspersen, J. G. and Aseervatham, V. (2012), ‘Affect, Heuristics and the Demand forInsurance’.URL: http: // ssrn. com/ abstract= 2028956

Louberge, H. (2000), Developments in Risk and Insurance Economics: the Past 25 Years,in G. Dionne, ed., ‘Handbook of insurance’, Kluwer Academic Publishers, Boston,pp. 3–33.

Michaelis, S. (2010), Die Haftung des Versicherungsmaklers gegenuber seinem Kunden:[nach dem neuen VVG], 1 edn, mbv, Mensch-und-Buch-Verl., Berlin.

Mossin, J. (1968), ‘Aspects of rational insurance purchasing’, Journal of Political Econ-omy 76(4), 553–568.

Schwarcz, D. (2010), ‘Insurance demand anomalies and regulation’, The Journal of Con-sumer Affairs 44(3), 557–577.

Shapira, Z. and Venezia, I. (2008), ‘On the preference for full-coverage policies: Whydo people buy too much insurance?’, Journal of economic psychology : research ineconomic psychology and behavioral economics 29(5), 747–761.

Zinnert, M. (2010), Versicherungsvermittler, 1. edn, Verl. Versicherungswirtschaft, Karl-sruhe.

Zweifel, P. and Eisen, R. (2012), Insurance Economics, Springer Berlin Heidelberg,Berlin and Heidelberg.

22

Working Paper Series in Economics

(recent issues)

No.298: Joachim Wagner: Extensive Margins of Imports and Profitability: First Evidence for Manufacturing Enterprises in Germany, May 2014

No.297: Joachim Wagner: Is Export Diversification good for Profitability? First Evidence for Manufacturing Enterprises in Germany, March 2014

No.296: Joachim Wagner: Exports and Firm Profitability: Quality matters!, March 2014

No.295: Joachim Wagner: What makes a high-quality exporter? Evidence from Germany, March 2014 [published in: Economics Bulletin 34 (2014), 2, 865-874]

No.294: Joachim Wagner: Credit constraints and margins of import: First evidence for German manufacturing enterprises, February 2014

No.293: Dirk Oberschachtsiek: Waiting to start a business venture. Empirical evidence on the determinants., February 2014

No.292: Joachim Wagner: Low-productive exporters are high-quality exporters. Evidence from Germany, February 2014 [published in: Economics Bulletin 34 (2014), 2, 745-756]

No.291: Institut für Volkswirtschaftslehre: Forschungsbericht 2013, Januar 2014

No.290: Stefan Baumgärtner, Moritz A. Drupp und Martin F. Quaas: Subsistence and substitutability in consumer preferences, December 2013

No.289: Dirk Oberschachtsiek: Human Capital Diversity and Entrepreneurship. Results from the regional individual skill dispersion nexus on self-employment activity., December 2013

No.288: Joachim Wagner and John P. Weche Gelübcke: Risk or Resilience? The Role of Trade Integration and Foreign Ownership for the Survival of German Enterprises during the Crisis 2008-2010, December 2013

No.287: Joachim Wagner: Credit constraints and exports: A survey of empirical studies using firm level data, December 2013

No.286: Toufic M. El Masri: Competition through Cooperation? The Case of the German Postal Market, October 2013

No.285: Toufic M. El Masri: Are New German Postal Providers Successful? Empirical Evidence Based on Unique Survey Data, October 2013

No.284: Andree Ehlert, Dirk Oberschachtsiek, and Stefan Prawda: Cost Containment and Managed Care: Evidence from German Macro Data, October 2013

No.283: Joachim Wagner and John P. Weche Gelübcke: Credit Constraints, Foreign Ownership, and Foreign Takeovers in Germany, September 2013

No.282: Joachim Wagner: Extensive margins of imports in The Great Import Recovery in Germany, 2009/2010, September 2013 [published in: Economics Bulletin 33 (2013), 4, 2732-2743]

No.281: Stefan Baumgärtner, Alexandra M. Klein, Denise Thiel, and Klara Winkler: Ramsey discounting of ecosystem services, August 2013

No.280: Antonia Arsova and Deniz Dilan Karamen Örsal: Likelihood-based panel cointegration test in the presence of a linear time trend and cross-sectional dependence, August 2013

No.279: Thomas Huth: Georg von Charasoff’s Theory of Value, Capital and Prices of Production, June 2013

No.278: Yama Temouri and Joachim Wagner: Do outliers and unobserved heterogeneity explain the exporter productivity premium? Evidence from France, Germany and the United Kingdom, June 2013 [published in: Economics Bulletin, 33 (2013), 3, 1931-1940]

No.277: Horst Raff and Joachim Wagner: Foreign Ownership and the Extensive Margins of Exports: Evidence for Manufacturing Enterprises in Germany, June 2013

No.276: Stephan Humpert: Gender Differences in Life Satisfaction and Social Participation, May 2013

No.275: Sören Enkelmann and Markus Leibrecht: Political Expenditure Cycles and Election Outcomes Evidence from Disaggregation of Public Expenditures by Economic Functions, May 2013

No.274: Sören Enkelmann: Government Popularity and the Economy First Evidence from German Micro Data, May 2013

No.273: Michael Berlemann, Soeren Enkelmann, and Torben Kuhlenkasper: Unraveling the Relationship between Presidential Approval and the Economy – A Multi-Dimensional Semi-Parametric Approach, May 2013

No.272: Michael Berlemann and Sören Enkelmann: The Economic Determinants of U.S. Presidential Approval – A Survey, May 2013

No.271: Soeren Enkelmann: Obama and the Macroeconomy Estimating Social Preferences Between Unemployment and Inflation, May 2013

No.270: Anja Köbrich León: Does Cultural Heritage affect Employment decisions – Empirical Evidence for Second Generation Immigrants in Germany, April 2013

No.269: Anja Köbrich León and Christian Pfeifer: Religious Activity, Risk Taking Preferences, and Financial Bahavior, April 2013

No.268: Anja Köbrich León: Religion and Economic Outcomes – Household Savings Bahavior in the USA, April 2013

No.267: John P. Weche Gelübcke and Isabella Wedl: Environmental Protection of Foreign Firms in Germany: Does the country of origin matter?, April 2013

No.266: Joachim Wagner: The Role of extensive margins of exports in The Great Export Recovery in Germany, 2009/2010, March 2013

No.265: John-Oliver Engler and Stefan Baumgärtner: Model choice and size distribution: a Bayequentist approach, February 2013

No.264: Chiara Franco and John P. Weche Gelübcke: The death of German firms: What role for foreign direct investment?, February 2013

No.263: Joachim Wagner: Are low-productive exporters marginal exporters? Evidence from Germany, February 2013 [published in Economics Bulletin 33 (2013), 1, 467-481]

No.262: Sanne Hiller, Philipp J. H. Schröder, and Allan Sørensen: Export market exit and firm survival: theory and first evidence, January 2013

No.261: Institut für Volkswirtschaftslehre: Forschungsbericht 2012, Januar 2013

No.260: Alexander Vogel and Joachim Wagner: The Impact of R&D Activities on Exports of German Business Services Enterprises : First Evidence from a continuous treatment approach, December 2012

No.259: Christian Pfeifer: Base Salaries, Bonus Payments, and Work Absence among Managers in a German Company, December 2012

No.258: Daniel Fackler, Claus Schnabel, and Joachim Wagner: Lingering illness or sudden death? Pre-exit employment developments in German establishments, December 2012

No.257: Horst Raff and Joachim Wagner: Productivity and the Product Scope of Multi-product Firms: A Test of Feenstra-Ma, December 2012 [published in: Economics Bulletin, 33 (2013), 1, 415-419]

No.256: Christian Pfeifer and Joachim Wagner: Is innovative firm behavior correlated with age and gender composition of the workforce? Evidence from a new type of data for German enterprises, December 2012

No.255: Maximilian Benner: Cluster Policy as a Development Strategy. Case Studies from the Middle East and North Africa, December 2012

No.254: Joachim Wagner und John P. Weche Gelübcke: Firmendatenbasiertes Benchmarking der Industrie und des Dienstleistungssektors in Niedersachsen – Methodisches Konzept und Anwendungen (Projektbericht), Dezember 2012

No.253: Joachim Wagner: The Great Export Recovery in German Manufacturing Industries, 2009/2010, November 2012 [published in: Review of Economics 64 (2013), 3, 325-340]

No.252: Joachim Wagner: Daten des IAB-Betriebspanels und Firmenpaneldaten aus Erhebungen der Amtlichen Statistik – substitutive oder komplementäre Inputs für die Empirische Wirtschaftsforschung?, Oktober 2012 [publiziert in: Journal of Labour Market Research 47 (2014), 1-2, 63-70]

No.251: Joachim Wagner: Credit constraints and exports: Evidence for German manufacturing enterprises, October 2012 [published in: Applied Economics 46 (2014), 3, 294-302]

No.250: Joachim Wagner: Productivity and the extensive margins of trade in German manufacturing firms: Evidence from a non-parametric test, September 2012 [published in: Economics Bulletin 32 (2012), 4, 3061-3070]

No.249: John P. Weche Gelübcke: Foreign and Domestic Takeovers in Germany: First Comparative Evidence on the Post-acquisition Target Performance using new Data, September 2012

No.248: Roland Olbrich, Martin Quaas, and Stefan Baumgärtner: Characterizing commercial cattle farms in Namibia: risk, management and sustainability, August 2012

No.247: Alexander Vogel and Joachim Wagner: Exports, R&D and Productivity in German Business Services Firms: A test of the Bustos-model, August 2012 [published in Empirical Economics Letters 12 (2013), 1]

No.246: Alexander Vogel and Joachim Wagner: Innovations and Exports of German Business Services Enterprises: First evidence from a new type of firm data, August 2012 [published in: Anna Maria Ferragina, Erol Taymaz and Kamil Yilmaz (Ed.), Innovation, Globalization and Firm Dynamics - Lessons for enterprise policy. Milton Park and New York: Routledge 2014, p. 137-160]

No.245: Stephan Humpert: Somewhere over the Rainbow: Sexual Orientation Discrimination in Germany, July 2012

No.244: Joachim Wagner: Exports, R&D and Productivity: A test of the Bustos-model with German enterprise data, June 2012 [published in: Economics Bulletin, 32 (2012), 3, 1942-1948]

(see www.leuphana.de/institute/ivwl/publikationen/working-papers.html for a complete list)

Leuphana Universität Lüneburg

Institut für Volkswirtschaftslehre

Postfach 2440

D-21314 Lüneburg

Tel.: ++49 4131 677 2321

email: brodt@leuphana.de

www.leuphana.de/institute/ivwl/publikationen/working-papers.html